pith. sign in

arxiv: 2605.23218 · v1 · pith:DCAOJRI3new · submitted 2026-05-22 · 💻 cs.AI

Foundation Protocol: A Coordination Layer for Agentic Society

Bang Liu , Yongfeng Gu , Jiayi Zhang , Zhaoyang Yu , Sirui Hong , Maojia Song , Xiaoqiang Wang , Mingyi Deng
show 21 more authors
Zijie Zhuang Ronghao Wang Mingzhe Cao Yutong Zhu Xingjian Li Yifan Wu Jianhao Ruan Yiran Peng Shuangrui Chen Jinlin Wang Yizhang Lin Dongjie Zhang Dekun Wu Chen Ma Lizi Liao Han Yu Jian Pei Heng Ji Qiang Yang Yuyu Luo Chenglin Wu
This is my paper

Pith reviewed 2026-05-25 04:42 UTC · model grok-4.3

classification 💻 cs.AI
keywords coordination layerautonomous agentsmulti-agent collaborationhuman-AI societyeconomic primitivesaccountabilitygraph protocolagentic systems
0
0 comments X

The pith

The Foundation Protocol unifies agents, humans, institutions, and resources in a graph-based coordination layer for collaboration and accountability.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper establishes the Foundation Protocol as a coordination layer for societies where autonomous agents interact with each other and with humans and institutions. It claims this layer uses a graph structure to connect these diverse entities, allowing them to organize, collaborate on events, exchange value through built-in economic tools, and maintain accountability via policies and audits. This approach is meant to work alongside existing systems rather than replace them, reducing the effort needed to integrate different agent technologies. A reader would care if they see the shift from individual agent performance to how groups of agents and people can reliably work together at scale.

Core claim

The Foundation Protocol is a graph-first coordination layer for an emerging human-AI society. It unifies heterogeneous entities including agents, tools, resources, humans, institutions, and organizations. It supports native multi-party organization and event-based collaboration, provides economic primitives for metering, receipts, and settlement, and treats policy, provenance, and audit as first-class concerns. The protocol is designed to wrap and bridge existing protocols to enable incremental adoption while keeping agency composable and accountability non-negotiable.

What carries the argument

The Foundation Protocol, a graph-first coordination layer that unifies heterogeneous entities and supports multi-party organization, economic exchange, and first-class policy and audit features.

Load-bearing premise

That a single graph-first protocol can be designed and incrementally adopted to solve coordination, economic exchange, and accountability problems across heterogeneous agents and institutions without introducing new integration or governance overhead that outweighs its benefits.

What would settle it

An experiment deploying the protocol across several independent agent platforms and measuring the actual coordination gains versus any added overhead in integration and governance.

read the original abstract

Autonomous agents are moving from tools into a layer of social infrastructure: they browse, purchase, deploy software, manage systems, and increasingly interact with one another. As these systems scale, the bottleneck shifts away from raw model capability toward coordination. Agents need to form reliable relationships, organize multi-agent work, exchange value, support an AI economy, and stay safe and accountable under real-world oversight. This paper introduces the Foundation Protocol (FP), a graph-first coordination layer for an emerging human-AI society. FP unifies heterogeneous entities, including agents, tools, resources, humans, institutions, and organizations, and supports native multi-party organization and event-based collaboration. It also provides economic primitives for metering, receipts, and settlement, and treats policy, provenance, and audit as first-class concerns. FP is designed to wrap and bridge existing protocols rather than replace them, enabling incremental adoption while reducing integration and governance overhead. The aim is to keep autonomous agency composable while keeping accountability non-negotiable, so that coordination itself can become shared infrastructure for a human-AI society that is open, pluralistic, and governable.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. The paper introduces the Foundation Protocol (FP), a graph-first coordination layer for an emerging human-AI society. FP is claimed to unify heterogeneous entities (agents, tools, resources, humans, institutions, organizations), support native multi-party organization and event-based collaboration, provide economic primitives for metering/receipts/settlement, treat policy/provenance/audit as first-class concerns, and wrap/bridge existing protocols to enable incremental adoption while reducing integration and governance overhead.

Significance. If realized with the claimed properties, FP could serve as shared infrastructure for reliable multi-agent coordination, economic exchange, and accountability in scaled agentic systems, addressing an emerging bottleneck beyond raw model capability. The proposal identifies a timely problem in human-AI societies but supplies no derivations, implementations, or evaluations to assess whether the unification and overhead-reduction claims hold.

major comments (2)
  1. [Abstract] Abstract: the claim that FP 'wraps and bridges existing protocols rather than replace them, enabling incremental adoption while reducing integration and governance overhead' is asserted without any interface specification, wrapping mechanism, or comparison showing net reduction in overhead; this is load-bearing for the incremental-adoption assertion.
  2. [Abstract] Abstract: no graph schema, node/edge typing for heterogeneous entities, event semantics, or metering primitives are provided, leaving the unification of agents/tools/humans/institutions and the support for multi-party organization/economic primitives unevaluable.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive feedback on our manuscript. We address the two major comments point by point below, indicating planned revisions where appropriate.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the claim that FP 'wraps and bridges existing protocols rather than replace them, enabling incremental adoption while reducing integration and governance overhead' is asserted without any interface specification, wrapping mechanism, or comparison showing net reduction in overhead; this is load-bearing for the incremental-adoption assertion.

    Authors: We agree that the incremental-adoption claim would be strengthened by concrete detail. The manuscript motivates the wrapping approach conceptually as a design principle for compatibility with existing agent, communication, and economic protocols, but does not supply interface specifications or overhead comparisons. In the revised manuscript we will add a dedicated subsection with a high-level wrapping architecture, one worked example of bridging to an existing protocol, and a qualitative discussion of integration overhead reduction. revision: yes

  2. Referee: [Abstract] Abstract: no graph schema, node/edge typing for heterogeneous entities, event semantics, or metering primitives are provided, leaving the unification of agents/tools/humans/institutions and the support for multi-party organization/economic primitives unevaluable.

    Authors: The current manuscript presents FP at the level of an architectural vision rather than a fully formalized specification. The graph model, heterogeneous node types, event-based collaboration, and economic primitives are described at a conceptual level. We acknowledge that the lack of explicit schemas limits evaluability. In the revision we will include a preliminary graph schema (with node and edge typing), definitions of core event semantics, and descriptions of the metering primitives in a new section or appendix. revision: yes

Circularity Check

0 steps flagged

No significant circularity; paper is a definitional proposal without derivation chain.

full rationale

The paper introduces the Foundation Protocol by definition, stating what FP unifies, supports, provides, and treats as first-class without any equations, predictions, fitted parameters, or load-bearing self-citations that reduce claims to inputs by construction. No mathematical steps, uniqueness theorems, or ansatzes are present to inspect for circularity. The abstract and description consist of aspirational feature lists rather than a derivation that could be shown equivalent to its inputs. This matches the default expectation for a protocol-design paper and receives the normal non-finding of score 0.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 1 invented entities

The proposal rests on the domain assumption that coordination will become the dominant bottleneck and that a graph-based unification layer can address it without new overhead. No free parameters or invented physical entities are introduced; the main invented entity is the protocol itself.

axioms (2)
  • domain assumption Coordination among autonomous agents will become the primary scaling bottleneck as agent numbers grow.
    Stated in the opening paragraph of the abstract as the premise that shifts the problem from model capability to coordination.
  • ad hoc to paper A single protocol can wrap and bridge existing protocols while reducing rather than increasing integration and governance overhead.
    Explicitly claimed in the abstract as the adoption strategy; no supporting argument or prior evidence is supplied.
invented entities (1)
  • Foundation Protocol (FP) no independent evidence
    purpose: To serve as the coordination layer that unifies entities and supplies collaboration, economic, and governance primitives.
    The protocol is the central new construct introduced by the paper; no independent falsifiable evidence for its properties is provided in the abstract.

pith-pipeline@v0.9.0 · 5826 in / 1624 out tokens · 19627 ms · 2026-05-25T04:42:10.635945+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

22 extracted references · 22 canonical work pages

  1. [1]

    https://a2a-protocol.org/latest/specification/

    A2A Protocol Specification (Overview). https://a2a-protocol.org/latest/specification/. Ac- cessed: 2026-03-01

    work page 2026

  2. [2]

    Accessed: 2026-03-01

    A2UI Specification v0.9.https://a2ui.org/specification/v0.9-a2ui/. Accessed: 2026-03-01

    work page 2026

  3. [3]

    Information Management: A Proposal

    Tim Berners-Lee. Information Management: A Proposal. Technical report, CERN, 1989

    work page 1989

  4. [4]

    A Next-Generation Smart Contract and Decentralized Application Platform.https: //ethereum.org/en/whitepaper/

    Vitalik Buterin. A Next-Generation Smart Contract and Decentralized Application Platform.https: //ethereum.org/en/whitepaper/. White paper. Accessed: 2026-03-01

    work page 2026

  5. [5]

    Some Simple Economics of AGI.arXiv preprint arXiv:2602.20946, 2026

    Christian Catalini, Xiang Hui, and Jane Wu. Some Simple Economics of AGI.arXiv preprint arXiv:2602.20946, 2026. Accessed: 2026-03-01

    work page arXiv 2026

  6. [6]

    Agent Network Protocol Technical White Paper.arXiv preprint arXiv:2508.00007, 2025

    Gaowei Chang, Eidan Lin, Chengxuan Yuan, Rizhao Cai, Binbin Chen, Xuan Xie, and Yin Zhang. Agent Network Protocol Technical White Paper.arXiv preprint arXiv:2508.00007, 2025. Accessed: 2026-03-01

    work page arXiv 2025

  7. [7]

    Accessed: 2026-03-01

    DIDComm Messaging Specification v2.0.https://identity.foundation/didcomm-messaging/spec/ v2.0/. Accessed: 2026-03-01

    work page 2026

  8. [8]

    Design Principles for Industrie 4.0 Scenarios: A Literature Review

    Mario Hermann, Tobias Pentek, and Boris Otto. Design Principles for Industrie 4.0 Scenarios: A Literature Review. In2016 49th Hawaii International Conference on System Sciences (HICSS), pages 3928–3937. IEEE, 2016

    work page 2016

  9. [9]

    Katz and Carl Shapiro

    Michael L. Katz and Carl Shapiro. Network Externalities, Competition, and Compatibility.The American Economic Review, 75(3):424–440, 1985

    work page 1985

  10. [10]

    Running OpenClaw safely: identity, iso- lation, and runtime risk

    Microsoft Defender Security Research Team. Running OpenClaw safely: identity, iso- lation, and runtime risk. https://www.microsoft.com/en-us/security/blog/2026/02/19/ running-openclaw-safely-identity-isolation-runtime-risk/ , 2026. Microsoft Security Blog. Ac- cessed: 2026-03-01

    work page 2026

  11. [11]

    https://modelcontextprotocol.io/ specification/2025-11-25

    Model Context Protocol Specification (2025-11-25). https://modelcontextprotocol.io/ specification/2025-11-25. Accessed: 2026-03-01

    work page 2025

  12. [12]

    Accessed: 2026-03-01

    Moltbook: The Front Page of the Agent Internet.https://www.moltbook.com/. Accessed: 2026-03-01

    work page 2026

  13. [13]

    Accessed: 2026-03-01

    OpenClaw: Personal AI Assistant.https://openclaw.ai/. Accessed: 2026-03-01

    work page 2026

  14. [14]

    https://github.com/openclaw/openclaw

    openclaw/openclaw (GitHub repository). https://github.com/openclaw/openclaw. Accessed: 2026- 03-01. 12

    work page 2026

  15. [15]

    What Is Web 2.0: Design Patterns and Business Models for the Next Generation of Software

    Tim O’Reilly. What Is Web 2.0: Design Patterns and Business Models for the Next Generation of Software. https://www.oreilly.com/pub/a/web2/archive/what-is-web-20.html, 2005. Published 2005-09-30. Accessed: 2026-03-01

    work page 2005

  16. [16]

    Edward Elgar Publishing, Cheltenham, UK, 2002

    Carlota Perez.Technological Revolutions and Financial Capital: The Dynamics of Bubbles and Golden Ages. Edward Elgar Publishing, Cheltenham, UK, 2002

    work page 2002

  17. [17]

    I Infiltrated Moltbook, the AI-Only Social Network Where Humans Aren’t Al- lowed

    Reece Rogers. I Infiltrated Moltbook, the AI-Only Social Network Where Humans Aren’t Al- lowed. https://www.wired.com/story/i-infiltrated-moltbook-ai-only-social-network/ , 2026. WIRED. Accessed: 2026-03-01

    work page 2026

  18. [18]

    Crown Business, New York, 2017

    Klaus Schwab.The Fourth Industrial Revolution. Crown Business, New York, 2017

    work page 2017

  19. [19]

    Leibo, Julian Jacobs, William A

    Nenad Tomašev, Matija Franklin, Joel Z. Leibo, Julian Jacobs, William A. Cunningham, Iason Gabriel, and Simon Osindero. Virtual Agent Economies.arXiv preprint arXiv:2509.10147, 2025. Accessed: 2026-03-01

    work page arXiv 2025

  20. [20]

    https://ucp.dev/latest/ specification/overview/

    Universal Commerce Protocol (UCP) Specification: Overview. https://ucp.dev/latest/ specification/overview/. Accessed: 2026-03-01

    work page 2026

  21. [21]

    A Survey of AI Agent Protocols.arXiv preprint arXiv:2504.16736, 2025

    Yingxuan Yang, Huacan Chai, Yuanyi Song, Siyuan Qi, Muning Wen, Ning Li, Junwei Liao, Haoyi Hu, Jianghao Lin, Gaowei Chang, Weiwen Liu, Ying Wen, Yong Yu, and Weinan Zhang. A Survey of AI Agent Protocols.arXiv preprint arXiv:2504.16736, 2025. Accessed: 2026-03-01

    work page arXiv 2025

  22. [22]

    PublicAffairs, 2019

    Shoshana Zuboff.The Age of Surveillance Capitalism: The Fight for a Human Future at the New Frontier of Power. PublicAffairs, 2019. 13 A Reference Implementation The Foundation Protocol is accompanied by an open-source reference stack: thefoundation-protocol runtime1 and the ai-link-net application-network implementation2. Both are non-normative. The main...

    work page 2019